Objective: The consequences of the recently proposed International Association of Diabetes in Pregnancy Study Group (IADPSG) criteria for gestational diabetes mellitus (GDM) in women with polycystic ovary syndrome (PCOS) are not known. We compared the prevalence rates and risk factors for GDM in PCOS women according to both the WHO and the modified IADPSG criteria. Design: Post hoc analyses from a randomized, multicenter study were used. Methods: Fasting and 2-h plasma glucose levels were measured using a 75 g oral glucose tolerance test. GDM was diagnosed according to both the WHO and the modified IADPSG criteria.Results: The prevalence rates of GDM according to the WHO and the modified IADPSG criteria were 9.2 and 15.0% at week 12, 18.7 and 18.7% at week 19, and 25.6 and 24.2% at week 32. Shorter stature and increased insulin levels were correlated with WHO-GDM, but not with modified IADPSG-GDM at weeks 12 and 19. Less weight gain in pregnancy predicted GDM according to both sets of criteria. GDM diagnosis was correlated with less maternal weight loss the first year post-partum. Conclusions: No difference was found in the prevalence of GDM between the two sets of criteria used. Less weight gain in pregnancy was associated with GDM, independent of the diagnostic criteria used. Reduced weight loss the first year post-partum in women with GDM raises the question of whether GDM diagnosis per se or the fact that these women lose less weight after pregnancy predicts later diabetes mellitus.